the effects of school resource officers on feelings of

The Effects of School

Resource Officers on

Feelings of School

Connectedness and Safety Word Count: 5443

I. Introduction

The implementation of School Resource Officers (SROs) in public schools, as well as

their effects on students, are becoming increasingly important. Although School Resource

Officers (SROs) have been implemented in U.S. high schools since the mid-1900s, school

crime was not yet a national concern at the time of the program’s conception. According to

sociologist Dr. Ben T. Brown, school crime was once thought to be a problem of exclusively

impoverished schools (Brown 2006). The shooting at Columbine High School challenged

this belief, as it was considered the deadliest school massacre in the nation’s history (Brown

2006). Events of school violence both in the past and present have heightened public concern

of school violence, reflected in the increased implementation of SROs as a prevention

measure.

Despite their prevalence, the effectiveness of SROs on improving school safety and

perceptions of safety have been widely debated. After the school shooting at Marjory

Stoneman Douglas High School in Parkland, Florida, many people, including President

Trump, have pushed for SROs in every school as violence prevention measures. However,

“The school in Parkland, Fla. had a school resource officer on duty during the shooting. The

Broward Sheriff said the officer remained outside in a defensive position at the time of the

shooting” (Corley 2018). The continuous debate surrounding school safety has evidently

raised questions as to the effectiveness of SROs after their failures in preventing attacks.

Furthermore, the use of SROs in public questions has called into question their further

effects on students. Research has focused on SROs effects on perceptions of safety and

feelings of school connectedness. A study conducted by Matthew T. Theriot and John G.

Orme, administered a school safety survey to a Southeastern United States school district

(Theriot, Orme 2016). It resulted that students who had more positive attitudes toward SROs

on campus felt safer (Theriot, Orme 2016). Although positive attitudes toward SROs have

been related to boosted perceptions of safety, they've also been found to decrease school

connectedness, or students' feelings of connection and belonging to a school. SROs have

been extensively researched to reveal their effects on students, shifting the focus outside of

their main function as security measures.

In order to contribute new insight to the current body of research, the present study will

use previously established methods in a student body that differs ethnically and

socioeconomically. By administering a self-report survey developed by credible researchers

and screening potentially exaggerated scores, the present study will serve to address the

question: How do perceptions of school resource officers on high school campuses affect

students’ perceptions of safety and school connectedness?

II. Literature Review

School Resource Officer Programs

Since the mid-1900s, school resource officer programs have been implemented in schools

across the U.S., employing officers (SROs) to patrol and investigate crimes on school properties.

However, the number of SROs increased dramatically during the 1990s, and by the year 2000,

nearly half of the U.S.’s police departments had assigned full time officers to serve as SROs at

schools (Brown 2006). According to a National Assessment of School Resource Officers by Jack

McDevitt and Jenn Panniello, the goal of the SRO program is to ensure safety (McDevitt,

Panniello 2005). SROs act as counselors and teach students within their school community in

such a way that helps prevent crime (McDevitt, Panniello 2005). While these SROs aim to deter

school crime, many studies have undermined their effectiveness. In a study by Arrick Jackson in

2002, a survey was administered to high school juniors and seniors, attending either an SRO or a

non-SRO school (Jackson 2002). This survey asked their opinions of the seriousness of

delinquency, perceptions of the SRO, their perceptions of being identified if committing

delinquent acts, interactions with SROs and their liking of the SRO (Jackson 2002). Although

they found that SROs had nearly no effect on any of the scales, students attending a high school

with an SRO believed they were less likely to be detected when involved in delinquent activity

(Jackson 2002). Jackson’s study found that SROs were mostly effective in deterring assault on

campus, but did not show any further effects on other crimes or the student perceptions of crime

(Jackson 2002). Studies like Jackson’s highlight instances in which SROs have not been

effective in deterring crime and ensuring safety in a school community.

SROs and Perceptions of Safety

Due to the increased implementation of SROs, studies have been conducted to explore

their effects on students and faculty, such as their perceptions of safety. In a study by Theriot and

Orme, a survey administered to a Southeastern United States school district showed that students

who had more positive attitudes toward SROs on campus felt safer (Theriot, Orme 2016).

Additionally, they concluded that student experiences with school violence, rather than

interactions with SROs, had a more significant impact on their feelings of safety (Theriot, Orme

2016). Another study by McDevitt and Panniello focused on the factors that affect both students’

comfort reporting crimes to SROs and their perceptions of safety in their schools (McDevitt,

Panniello 2005). Through a survey administered to middle and high school students, they found

that knowing the SROs’ name as well as having conversations with them could positively affect

whether the student felt comfortable with reporting crimes to their SRO (McDevitt, Panniello

2005). In this study, perceptions of safety were most impacted by the level of neighborhood

crime, past victimization at school, their comfort reporting crimes, having a positive opinion of

SRO, and gender (McDevitt, Paniello 2005). Since there have been various interpretations of the

effects SROs have on perceptions of safety, it is imperative that further research is conducted to

make these effects clearer.

Student Perceptions of School Security Measures and Safety

Understanding the factors that construct one’s perceptions of safety provide insight into

how SROs might affect a students’ perceptions of safety. One of these factors is the presence of

security measures. In a study by Bosworth and colleagues in 2011, when randomly selected

students and teachers were asked “What makes a school safe”, they replied within 3 categories:

physical characteristics and safety features, organization and school discipline, and school

staffing and relationships that enhance feelings of safety with a sense of caring and community

(Bosworth et al. 2011). From these three, students ranked tangible security equipment higher

than school staffing as a contributor to feelings of safety (Bosworth et al. 2011). Both students

and faculty found that school organization and discipline, as well as school staffing and

relationships, made them feel safer at school (Bosworth et al. 2011). This study showed that both

students and staff felt safest in schools where they perceived the adults to be caring and helpful,

and that rules were clear, consistent, well communicated, and consistently applied (Bosworth et

al. 2011). These findings suggest that adults and students share factors that construct their

perceptions of safety. Another study by Ben Brown and W.M. Reed Benedict in 2005 focused on

student perceptions of school security measures and the effect on their feelings of safety (Brown,

Benedict 2005). A survey administered to students of the Brownsville Independent School

District showed that while students are supportive of many of the safety measures, those

employing “hard control” have a limited impact on school crime overall (Brown,Benedict 2005).

These hard control methods, such as video surveillance, are similar to the physical security

measures studied by Bosworth and colleagues to which they referred to as physical

characteristics and safety features. However, there are contradictory results between the two

studies since the students surveyed by Bosworth considered these measures to be more important

to their feelings of safety, but the students surveyed by Brown and Benedict found the measures

to be ineffectual in deterring crime. While security measures are an essential part of student

perceptions of safety overall, the conflicting opinions lead researchers to believe there are other

relevant factors.

The study “Beyond Guns, Drugs and Gangs” by Skiba and colleagues in 2004 addresses

these relevant factors. The researchers surveyed students, staff, and parents about school safety

(Skiba et al. 2004). The most important factors of their feelings of safety for all three groups

were the connection or climate of the school, incivility and disruption, personal safety, and

delinquency or major safety (Skiba et al. 2004). This study showed that school connection and

climate are more important than factors of delinquency or physical safety measures in predicting

students’ overall feelings about school safety in the locations studied by the researchers (Skiba et

al. 2004). Additional studies have also indicated that experiences with different forms of

aggression contribute to students’ perceptions of their school climate as well as their feelings of

safety at school. A study conducted by Sara Goldstein and colleagues called “Relational

Aggression at School: Associations with School Safety and Social Climate” surveyed middle

school and high school students in metropolitan Detroit provided insight into how students’

experiences at school affect their perceptions of safety (Goldstein et al. 2007) . Their survey

results indicated that high exposure to relational (verbal) aggression made students feel less safe

at school and have negative overall experiences at school, impacting their perceptions of their

social climate negatively (Goldstein et al. 2007). Since school climate and connection have been

found to have an impact on student perceptions of safety, the impact SROS have on those two

factors must also be addressed.

SROs and Feelings of School Connectedness

Another focus taken on the effects of SROs are their influence on the school

connectedness perceived by students. In a study by Matthew T. Theriot in 2016, the researcher

administered a survey to a school district in Southeastern United States about school safety

(Theriot 2016). The survey collected the amount of times students reported interacting with

SROs, and the students’ responses to questions about school connectedness and climate. These

questions were compiled into distinct scales to construct scale scores (Theriot 2016). The higher

the participants scored on either the school connectedness or SRO perceptions scales, the more

negative feelings of school connectedness or more positive opinions of the SRO were observed

(Theriot 2016). The survey showed that increased student interaction with SROs caused more

positive attitudes toward the SROs, but lower school connectedness. This was explained by

Theriot to possibly be a result of the increased interactions with SROs drawing attention to

school crime, which can generate worry or violence among students, and contribute to the

students feeling less connected and comfortable at school (Theriot 2016). By considering how

SROs affect school connectedness, there is a better understanding of how they affect students’

perceptions of safety overall, considering the conclusion the study by Skiba and colleagues that

identified it as a relevant factor to how safe students feel at school.

Validity of Self Report Survey Method

The majority of studies on student perceptions of safety have used self-report surveys to

identify the effects of SROs on students and faculty. Due to the reliance on this method, many

have assessed the validity of self-report surveys to gauge perceptions of safety in response to

SROs. In a study by Cornell and Loper, it was demonstrated that by screening surveys for

incomplete or careless responses, surveys can be essential in gaining insight into the nature and

extent of high-risk behavior in schools, and effectively help school psychologists and other

educators in developing violence intervention and prevention programs (Cornell, Loper 1998).

These methods established by Cornell and Loper were used in Theriot’s various studies on

school resource officers, including in his work in conjunction with John G. Orme. This was done

to exclude careless reporting or surveys that showed high levels of victimization. While Theriot

and Orme explicitly used these screening methods to ensure the validity of their survey

responses, much of the research about student perceptions of safety and SROs does not. For

example, the study “Police-School Officers and students’ perceptions of police and offending”

by Arrick Jackson implements the self-report survey method to measure student perception of

delinquency and school-resource officers or police, depending on whether the school they

focused on used an SRO. The survey data indicated that increased interaction with SROs resulted

in more positive opinions of the students’ SRO, but it did not change their opinions of police

outside of their school (Jackson 2002). While this survey was consistent with the results of the

study done by Theriot in 2016, there was no screening process implemented during data analysis,

which is essential in excluding careless responses and improving the validity of the self-report

survey method. This lack of screening proves to be missing in much of the research in this area.

Hypothesis

The mission of this research is to explore the relationships between school resource

officers and high school students’ perceptions of safety and school connectedness. The

researcher of this study hypothesizes that in an environment that differs ethnically and

socioeconomically than previous studies, school safety survey results will support previous

studies that found SROs to affect perceptions of safety positively and student’s feelings of school

connectedness negatively. This would be conveyed through increased feelings of safety but

decreased levels of school connectedness in response to more interactions with or more positive

opinions of the SRO.

III. Methods

A. Population: The school included the present study, GHC, is a large

sized co-ed public high school in a suburban setting. There are 4,480

attending students, with an ethnically diverse population, outlined in

Table 1. GHC’s population contrasts with this study’s foundational

sources due to differences in regional settings and ethnic demographics.

These differences may prove to create a difference between this the

results of this study’s population and of its foundational sources’

population.

B. Survey Alignment: In order to align with previous methods implemented in studying

student perceptions of SROs and safety, a researcher expert was contacted. Matthew T.

Theriot and John G. Orme, professors of the College of Social Work at the University of

Tennessee, have previously researched student perceptions of SROs through survey

distribution. In order to closely align the present research to previously established

Race/

Ethnicity

Percentage

(%)

Hispanic 40

White 25

Asian 18

Filipino 9

African

American 4

American

Indian/Alaska

Native <1Pacific

Islander <1

Declined To

State <2

Source: GHC 2017-18 School Profile

Table 1: GHC Ethnic

Demographics

methods, Theriot was contacted through email and he provided the survey used in his and

Orme’s study. This survey was then included in the present study to be distributed after a

stratified random sample of GHC student was selected. From this survey instrument, the

questions making up each scale regarding school connectedness, school safety

perceptions, violence experienced at school are described in Table 2.

TABLE 2: Instrument Table

Questions Measurement Scale Source

Demographic Questions Assorted

What gate did you enter the school from the morning you were contacted?

Flagpole/Zelzah, Zelzah Teacher Parking Lot, Service Road

(Boy's PE side), Hiwatha Parking Lot, Girls PE Side Service

Road, J Gate, Kingsbury (main entrance) self-defined

What is your gender? Male, Female self-defined

What grade are you in at schoool? Freshman, Sophmore, Junior, Senior self-defined

How many years have you been a student at this school? 1 (this is my first year), 2, 3, 4 ,5 or more Theriot, Orme 2016

How many good friends do you have at school?

None, I have one good friend at school, I have 2-3 good friends

at school, I have 4-5 good friends at school, I have 6 or more

good friends at school

School Connectedness Scale Statements Likert Scale

I am learning a lot at this school.

5 = strongly agree; 3 =agree, neutral; and 1=disagree, strongly

disagree Theriot, Orme 2016

School rules seem fair to me.


disagree

Teachers work hard to make every student successful.


disagree

I feel that I belong at this school.


disagree

I do not try hard in school.


disagree

In general, I like school.


disagree

I don’t care what teachers think of me.


disagree

I am proud of this school.


disagree

I do not feel that I can tell a teacher, principal, or other adult at school if I have a problem.


disagree

Students enjoy learning here.


disagree

Student School Safety Perceptions Statements Scale Likert Scale

Overall, I feel safe at school. 1=strongly disagree; 3=neutral; and 5=strongly agree Theriot, Orme 2016

I feel safe in the school hallways. 1=strongly disagree; 3=neutral; and 5=strongly agree

I feel safe in the cafeteria. 1=strongly disagree; 3=neutral; and 5=strongly agreeI feel safe going to school in the morning and coming home from school in the

afternoon. 1=strongly disagree; 3=neutral; and 5=strongly agree

I feel safe in the bathrooms at school. 1=strongly disagree; 3=neutral; and 5=strongly agree

I feel safe in my classrooms. 1=strongly disagree; 3=neutral; and 5=strongly agree

In general, I do not trust the police. 1=strongly disagree; 3=neutral; and 5=strongly agree

Having a police officer at my school everyday would make me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree

Having metal detectors at my school would make me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree

Having surprise locker checks at my school would help me feel safer. 1=strongly disagree; 3=neutral; and 5=strongly agree

In general, I like the police. 1=strongly disagree; 3=neutral; and 5=strongly agree

Gangs are a big problem at my school. 1=strongly disagree; 3=neutral; and 5=strongly agree

Student Experience with Violence at School Likert Scale

During this school year, how often have you been teased, called names, or insulted

by another student while at school?

1=never, 2=once or twice, 3=about once a month, 4=2–3 times a

month, 5= about once a week, and 6= several times Theriot, Orme 2016

During this school year, how often have you had a physical fight with another

student while at school?


month, 5= about once a week, and 6= several times

During this school year, how often have you had an argument with another student

while at school?



During this school year, how often have you been bullied by another student while

at school?



During this school year, how often have you been hit, kicked, grabbed, or otherwise

physically grabbed, or otherwise physically harmed by a boyfriend, girlfriend, or

anyone that you were dating?



During this school year, how often have you been teased, called names, insulted, or

attacked at school because of your skin color, race, religion, or where your family

came from?



During this school year, how often has another student threatened to physically

harm you while at school?



School Violence Students Have Seen Likert Scale

During this school year, how many times have you seen students smoke cigarettes or use

tobacco at school?


month, 5= about once a week, and 6= several times Theriot, Orme 2016

During this school year, how many times have you seen a knife at school (not including a

cafeteria knife)?



During this school year, how many times have you seen a gun at school (not including police

officers)?



During this school year, how many times have you seen students with drugs or alcohol at

school?



During this school year, how many times have you seen a student tease

or insult another student (including name-calling)?



During this school year, how many times have you seen a student bully another student?



During this school year, how many times have you heard a student threaten to physically harm

another student?



During this school year, how many times have you seen violence between two students who are

dating each other?



During this school year, how many times have you seen two or more students have a loud

argument?



During this school year, how many times have you seen two or more students get in a physical

fight?



During this school year, how many times have you seen a student tease, insult, or attack

another student because of skin color, religion, or where they came from?



Attitudes About SROs Scale Statements Likert Scale

I like having a police officer at my school everyday.


disagree Theriot, Orme 2016

I feel safer with the police officer at my school.


disagree

The police officer at my school does not treat all students fairly.


disagree

In general, the police officer at my school does a good job.


disagree

The police officer at my school has a good relationship with the students.

= strongly agree; 3 =agree, neutral; and 1=disagree, strongly

disagree

The police officer l does a good job of stopping violence at my school.


disagree

The police officer does a good job of stopping students from selling or using drugs or alcohol at

school.

5= strongly agree; 3 =agree, neutral; and 1=disagree, strongly

disagree

The police officer at my school does not make me feel better about the police in general.

5= strongly agree; 3 =agree, neutral; and 1=disagree, strongly

disagree

The police officer at school will help students who need help


disagree

Most students at my school don't like the police officer.


disagree

Number of Interactions with SRO Assorted

During this school year, how many times have you interacted with the police officer at your

school (including things like talking to him or her, asking for or getting help from the police

officer, and getting in trouble)? None, 1-2 times, 3-4 times, 5-6 times, 7 or more times Theriot, Orme 2016

During this school year, has the police officer at your school led a class or given a presentation

that you attended? Yes , No

Source: My December 2018 GHC Study

C. Sample Selection: In order to make each student equally likely to be chosen to

participate in the study, a stratified random sample was done every day before school for

3 weeks. Each of the seven gates on campus has a certain percentage of students who

come through it. In order to perform a stratified random sample, students were randomly

chosen and asked to participate in the study. The gate strata, and the percentage of

students included are outlined in Table 3. With the help of two business statistics

students, students coming into school through each gate were asked if they wanted to

participate in the study. All GHC students have access to a Chromebook and Gmail

account, providing an easily accessible method of distributing the survey. Through this

process, 125 students were able to participate in the present study. While there was a

sample of 125 students from all of the gates, a further random sample of 100 participants

out of the 125 was chosen using Excel technology in order to facilitate the data

processing and assure the gate proportions were met for each stratum.

School EntrancePercentage of Students

(%)

Flagpole (Zelzah) 28

Kingsbury 24

Zelzah Teacher

Parking Lot22

Service Road

West11

Hiawatha Lot 7

Service Road East 5

J-Gate 3


Table 3: Stratified Random Sample

D. Implementation: The survey implemented in this study was mechanically distributed

through school-issued Chromebooks and Google Forms. Each time a participant was

identified at the gate, the survey was sent to them digitally through an email with a

Google Forms link. Since each student has access to the technology implemented, every

student in the population was equally likely to participate. The data from each survey was

processed through Excel Toolpak functions such as histogram constructions or univariate

data calculations.

E. Survey Screening Alignment: A survey screening process was implemented to validate

the results, removing any incomplete or exaggerated surveys. This was done in alignment

with the study by Cornell and Loper demonstrated that by screening surveys for

incomplete or careless responses, surveys can be essential in gaining insight into the

nature and extent of high-risk behavior in schools (Cornell and Loper 1998). Similarly,

the study “Assessing the Reliability and Validity of Student Self-Reports of Campus

Violence” by Jennifer Rosenblatt and Michael Furlong indicated that screening for

incomplete responses and designing procedures that exclude illegitimate answers allow a

more accurate study of school safety and campus violence (Rosenblatt and Furlong

1997). This screening process, implemented through the Google Surveys technology, is

integral to the validity of the self-report survey used in this study, due to the many of the

questions’ subject matter of school violence. It is possible that some respondents may

have carelessly filled out the survey and inaccurately portrayed the sample population, so

by screening the results, the present study has aligned itself to previous methods of

survey distribution.

IV. Findings

A. Histograms

62%

38%

Female Male


What is your gender?

26%

21%

26% 27%

Freshman Sophomore Junior Senior


What grade are you in at school?

26%24% 24%

26%

1 (this is my firstyear at this school)

2 years 3 years 4 years


How many years have you been a student at this school?

0 1

17

3943

None I have 1 goodfriend atschool

I have 2-3good friends at

school

I have 4-5good friends at

school

I have 6 ormore goodfriends at

schoolSource: My December 2018 GHC Study

How many good friends do you have at school?

23

5

1

2019

1716

12

2 20 0 0 0

1

16 18 20 21 22 24 26 28 30 32 34 36 38 40 42 44Source: My December 2018 GHC Study

School Connectedness Scores

01

0 0

21

0

4

21

6

9

3

89

14

17

32

6

23

7

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30Source: My December 2018 GHC Study

Safety Perceptions Scale Scores

B. Descriptive Statistics

27

19 19

10

6

3 42 1 1 1 1 0

20 1

3

0 0 0 0 0 0 0

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30


Experienced Violence Scale Scores

5

4

5

7

9 9

7

0

8

4 4

2

3

7 7

1 1

3

0

2

0

2

1

2

0

1 1 1

0

1

0 0 0 0 0 0

1

0

1 1

11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49


Witnessed Violence Scale Scores

2

6

15

32

1922

2 2

20 22 24 26 28 30 32 34


SRO Perceptions Scale Score

94

6 0 0 0

None 1-2 times 3-4 times 5-6 times 7 or moretimes


During this school year, how many times have you interacted with the police officer at your

school (including things like talking to him or her, asking for or getting help from the police officer,

and getting in trouble)?

5

95

Yes NoSource: My December 2018 GHC Study

During this school year, has the police officer at your school led a class or given a presentation

that you attended?

C. Correlations

D. Regressions

Survey Scale Possible Range Mean score Standard Deviation

School Connectedness 10 to 50 25.29 0.417

Safety Perceptions 6 to 30 22.32 4.37

Experienced Violence 7 to 42 10.11 4.32

Witnessed Violence 7 to 42 20.95 8.35

SRO Perceptions 10 to 50 26.88 2.8

SC Score SP Score EV Score WV Score SROP Score

SC Score 1.00

SP Score 0.43 1.00

EV Score -0.29 -0.28 1.00

WV Score -0.18 -0.23 0.65 1.00

SROP Score 0.19 0.20 -0.09 -0.09 1.00

slight positive slight negative

strong positive

Interactions Class/Presentation SC Score SP Score SROP Score

Interactions 1.00

Class/Presentation -0.06 1.00

SC Score -0.31 0.02 1.00

SP Score -0.01 -0.05 0.43 1.00

SROP Score -0.02 0.12 0.19 0.20 1.00

slight negative slight positive

V. Analysis of Findings

The foundational study by Theriot and Orme utilized Latent Class Analysis (LCA) and

logistic modeling regression. Regression to analyze the survey results. The LCA serve to identify

discrete subpopulations, or latent classes, of similar cases. The logistic modeling regression

implemented the latent classes identified in the LCA in bivariate and multivariate regression. In

order to best align the present study’s analysis method to those of their foundational studies, the

researcher of the present study utilized the descriptive statistics, correlation, and multivariate

logistic regression functions available on Excel technology.

Comparing Descriptive Statistics and Overall Trends

Due to the manner in which this study has aligned itself with the study done by Theriot

and Orme, the survey data descriptive statistics and overall trends were important to highlight

any differences between the two populations. The survey was broken into demographics, then

into multiple scales addressing school connectedness, safety perceptions, violence experienced,

violence witnessed, and SRO perceptions. The survey was concluded with two questions

Safety Perceptions Score Regression

CoefficientsStandard Error t Stat P-value

Intercept 10.57 4.51 2.34 0.02

SC Score 0.37 0.10 3.78 0.00

EV Score -0.21 0.10 -2.13 0.04

SROP Score 0.16 0.14 1.16 0.25

School Connectedness Score Regression

Coefficients Standard Error t Stat P-value

Intercept 15.08 4.22 3.58 0.00

SP Score 0.35 0.09 3.78 0.00

EV Score -0.16 0.09 -1.65 0.10

SROP Score 0.15 0.14 1.09 0.28

indicating levels of interaction with the SRO. Each demographic question and stand-alone scale

was run through descriptive statistics with Excel technology and examined for overall trends.

Demographics

The first questions addressed the participants’ demographics, such as gender, grade, years

spent at this particular school, and number of good friends at school. The gender demographics

were 62 percent female and 38 percent male. For grade level, the proportions of each group were

close to equal, where there were 26% freshmen, 21% sophomores, 26% juniors, and 27%

seniors. This remained true for the spread for years spent at school since it was nearly uniform,

meaning that each group was nearly equal. For the number of good friends at school, more than

half of the respondents (72%) have at least 4 good friends at school. Only small percentage of

students (18%) have less than 4 good friends at school. Compared to the Theriot and Orme study

in which 1,744 students had 2 or more good friends at school (89%), the present study had 99

students with 2 or more good friends at school (99%) (Theriot, Orme 2016).

School Connectedness

The stand-alone scales were also run

through descriptive statistics with Excel

technology to identify any trends. The first survey

scale was after the demographic questions,

addressing the participants’ feelings of school

connectedness with 10 questions. It asked

questions about how fair they believed their school rules to be, work ethic, school pride, sense of

belonging, liking of school, whether the students are learning a lot and their willingness to report

23

5

1

2019

1716

12

2 20 0 0 0

1

16 18 20 21 22 24 26 28 30 32 34 36 38 40 42 44Source: My December 2018 GHC Study

School Connectedness Scores

problems to school staff. Each question within this scale had Likert scale responses where 5 =

strongly agree; 3 =agree, neutral; and 1=disagree, strongly disagree. To create the overall score,

each response was simply summed to one score. The school connectedness scale had a possible

range of 10 to 50. The descriptive statistics showed a mean school connectedness score of 25.29

with a standard deviation of .417. This is much lower than the mean school connectedness score

in the Theriot and Orme study of 33.2 with a standard deviation 6.5 (Theriot, Orme 2016). This

not only indicated lower levels of school connectedness in the participants of the present study,

but also a much less variable sample of school connectedness scores due to the lower standard

deviation. Additionally, the scale histogram was skewed to the right with higher proportions of

scores placed on the lower end of the range. This further indicates that the sample population has

lower levels of school connectedness.

Perceptions of Safety

The second survey scale, comprised of

6 questions, was meant to measure perceptions

of safety. Using the same 5 point Likert scale

as the last survey scale, students were asked if

they felt safe in various locations. This

included at school overall, in school hallways,

in the cafeteria, going to and from school, in school bathrooms, and their classrooms. The scale

score was summed the same as the last scale. The perceptions of safety scores had a possible

range of 6 to 30. The descriptive statistics showed a mean safety perceptions score of 22.32 and

standard deviation of .4366. Since the possible scores ranged from 6 to 30, the mean safety

perceptions score indicates higher feelings of safety within this sample population. The scale

01

0 0

21

0

4

21

6

9

3

89

14

17

32

6

23

7

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30Source: My December 2018 GHC Study

Safety Perceptions Scale Scores

histogram is also skewed to the left, with higher proportions of scores on the higher end of the

score range. This also supports the conclusion that this sample reported higher feelings of safety.

Experienced Violence

The third survey scale, comprised of 7

questions, measures the level of experienced

violence of the survey respondent. It asked the

participant questions of various types of violence

they may have personally experienced while at

school. This includes teasing, physical fights,

arguments, bullying, physical harm from a partner, racially based harassment, or physical threats.

The scores had a possible range from 7 to 40, in which higher score indicated higher levels of

experienced violence while at school. The sample population had a mean experienced violence

score of 10.11 with a standard deviation of 4.32. Compared to the sample population of the

Theriot and Orme study, which had a mean experienced violence scores of 12.8 and standard

deviation of 5.8, the present study’s sample indicated lower scores and less variability (Theriot,

Orme 2016). While the present study’s sample population had a lower mean experienced

violence score than the Theriot and Orme study, 12.8 is still within one standard deviation from

10.11, so it is not statistically significant. The scale histogram was dramatically skewed to the

right, with most scores on the lower end of the range. This is another indication that the current

study reported lower levels of experienced violence.

Witnessed Violence

27

19 19

10

6

3 42 1 1 1 1 0

20 1

3

0 0 0 0 0 0 0

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30


Experienced Violence Scale Scores

The fourth survey section addressed

any witnessed violence. It included the

same types of violence as the last scale, but

it instead asked the participant whether

they had seen it happen to another student,

rather than to themselves. The scores

ranged from 7 to 40. The survey sample had a mean witnessed violence score of 20.95 with a

standard deviation of 8.35. While the Theriot and Orme study provided no descriptive statistics

for this scale, the mean witnessed violence score of the present study indicates that its population

had lower levels of witnessed violence. This is due to the mean score, 20.95, being lower than

the median score of the possible range (23.5). Overall, the scale histogram shows a moderate

skewness to the right. This suggests that similarly to the experienced violence scale, the current

study’s sample population reported lower levels of witnessed violence.

Opinions of SRO

This scale asked the participant 10

questions addressing their perceptions of their

SRO. This includes whether they liked having

the officer at school, felt safer with the SRO at

school, and whether they believed the officer

was helpful to students. Additionally, students

were asked if they believed the SRO treated all students fairly, does a good job, makes them feel

better about police, has a good relationship with students, and that other students like the officer.

They were also asked whether they felt their SRO does a good job of stopping violence and

5

4

5

7

9 9

7

0

8

4 4

2

3

7 7

1 1

3

0

2

0

2

1

2

0

1 1 1

0

1

0 0 0 0 0 0

1

0

1 1

11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49


Witnessed Violence Scale Scores

2

6

15

32

1922

2 2

20 22 24 26 28 30 32 34


SRO Perceptions Scale Score

preventing crime (sale of drugs or alcohol at school). There was a possible range of 10 to 50

points on this scale. The current study had a mean SRO perception score of 26.88 and a standard

deviation of 2.8. Compared to the Theriot and Orme study with a mean score of 33.8 and

standard deviation of 6.5, the present study’s sample population had less positive opinions of

their SRO (Theriot, Orme 2016). The scale histogram points to the same conclusion given its

higher concentration of scores at or under 30 (the median of the possible score range). Overall,

the sample population has reported less positive opinions of its SRO on campus.

Scale Correlations

Many relationships can be identified between the measured scales of school

connectedness, experienced violence, witnessed violence, and SRO perceptions. Previous

research, such as the study by Skiba and colleagues has indicated that school connection and

climate are more important than factors of delinquency or physical safety measures (such as

SROs) in predicting students’ overall feelings about school safety (Skiba et al. 2004).

Additionally, the study done by Theriot and Orme demonstrated that while students with more

positive attitudes toward SROs on campus felt safer, experiences with school violence, rather

than their interactions with SROs, had a more significant impact on their feelings of safety

(Theriot, Orme 2016). The study conducted by McDevitt and Panniello also found that

perceptions of safety were most impacted by the level of neighborhood crime, past victimization,

comfort reporting crimes, having a positive opinion of SRO, and gender (McDevitt, Panniello

2005). In order to identify the possible relationships established by previous researchers, various

correlations between the multiple scales were run using Excel technology.

Safety Perceptions

The correlations demonstrated that relationships similar to the ones found in previous

research were also present in this study’s sample population. For example, there was a slight

positive correlation of +.43 between the School Connectedness (SC) and Safety Perceptions (SP)

scores. Meanwhile, there was only a correlation of +.20 between the SRO Perceptions (SROP)

scores and the SP Scores. This is consistent with the results of the study done by Skiba and

colleagues, which also indicated that school climate and connection was more important than

security measures in predicting feelings of school safety (Skiba et al. 2004). This is also

supported by how there was only a slight negative correlation of -.28 between Experienced

Violence (EV) and SP scores and only a slight negative correlation of -.23 between Witnessed

Violence (EV) and SP scores. Just as in Skiba and colleagues’ study, school climate and

connection was more important in predicting safety perceptions than student delinquency.

Overall, SC scores had the strongest correlation with the SP scores.

Interactions and Opinions of SRO

SC Score SP Score EV Score WV Score SROP Score

SC Score 1.00

SP Score 0.43 1.00

EV Score -0.29 -0.28 1.00

WV Score -0.18 -0.23 0.65 1.00

SROP Score 0.19 0.20 -0.09 -0.09 1.00

slight positive slight negative

strong positive

Interactions Class/Presentation SC Score SP Score SROP Score

Interactions 1.00

Class/Presentation -0.06 1.00

SC Score -0.31 0.02 1.00

SP Score -0.01 -0.05 0.43 1.00

SROP Score -0.02 0.12 0.19 0.20 1.00

slight negative slight positive

In order to identify a relationship between interactions with the SROs and opinions of the

SROs, correlations were run between the two questions quantifying the student’s number of

interactions with the SRO and the SROP scale. Contrary to the results of the Theriot study of

SROs and school connectedness in 2016, the correlations indicated no significant relationship

between the number of interactions between students and SROs and more positive opinions of

the SROs (Theriot 2016). This is due to the correlation of -.02 between SROP and Interactions

and the correlation of +.12 between SROP and Classes/Presentations. However, the number of

interactions had a slight negative correlation of -.31 with the SC score, which is consistent with

the same Theriot study which indicated that increased interactions with SROs was related to

lower levels of school connectedness (Theriot 2016).

Multivariate Logistic Regression

Due to the only slight correlations between opinions of SROs and perceptions of safety

and school connectedness, further data analysis was done to identify any other significant

relationships. Two multivariate regressions were performed for the Safety Perceptions Scores

and the School Connectedness Scores.

Predicting Safety Perceptions

The most significant variable

in predicting Safety Perception (SP)

Scores was the SC Scores. The

multivariate regression resulted in the

SC scores having a coefficient .37 and p-value of 0% when predicting SP Scores. The coefficient

is the biggest out of the 3 variables, and it is most statistically significant due to its low p-value.

Safety Perceptions Score Regression

CoefficientsStandard Error t Stat P-value

Intercept 10.57 4.51 2.34 0.02

SC Score 0.37 0.10 3.78 0.00

EV Score -0.21 0.10 -2.13 0.04

SROP Score 0.16 0.14 1.16 0.25

The coefficient of the EV is much lower, with a slightly higher p-value of .04. While the EV

scores are still a statistically significant variable, its lower coefficient indicates it is less

important in predicting SP Scores. Once again, the data analysis is consistent with Skiba and

colleagues’ study, which concluded that school climate and connection was more important in

predicting safety perceptions than student delinquency, or EV scores in the present study (Skiba

et al. 2004). It is also consistent with the results of Goldstein and colleagues study where high

exposure to relational (verbal) aggression, which was included in the EV scale, made students

feel less safe at school (Goldstein et al. 2007). Additionally, the coefficient of the SROP was

notably lower than of SC, and its p-value of 25% indicated the variable was statistically

insignificant. This is consistent with the results of Theriot and Orme’s study that concluded

students’ experiences with school violence, rather than their interactions with SROs, had a more

significant impact on their feelings of safety (Theriot, Orme 2016).

Predicting School Connectedness

The most significant variable

in predicting SC scores was the SP

Scores. It had a coefficient of .35

and p-value of 0%. However, the

least impactful variable was the SROP Scores, since its coefficient of SROP was .15 and its p-

value was 28%, indicating it was statistically insignificant. This is contrary to the results of

Theriot’s study in 2016 concluded that SROs negatively affected school connectedness (Theriot

2016). The second most impactful variable was the EV Scores, with a coefficient of -.16 and p-

value of 10%. While the p-value was too big to label the variable statistically significant, it’s

negative impact on SC scores is consistent with the results of Goldstein and colleagues study

School Connectedness Score Regression

Coefficients Standard Error t Stat P-value

Intercept 15.08 4.22 3.58 0.00

SP Score 0.35 0.09 3.78 0.00

EV Score -0.16 0.09 -1.65 0.10

SROP Score 0.15 0.14 1.09 0.28

where high exposure to relational (verbal) aggression, which was included in the EV scale,

impacting students’ perceptions of their social climate negatively (Goldstein et al. 2007).

VI. Conclusion

The present study has provided further insight into the effects of SROs on safety

perceptions and feelings of school connectedness. The overall trends, correlations, and

multivariate regressions indicated that while SRO perceptions do impact feelings of safety and

school connectedness, experienced violence is instead the most impactful variable. The effects

on experienced violence is supported by the study conducted by Goldstein and Boyd’s study

where high exposure to relational (verbal) aggression, which was included in the EV scale,

impacting students’ perceptions of their social climate negatively and feel less safe at school.

Despite this, the number of interactions with the SRO still had an impact on feelings of school

connectedness due to its negative correlation. Overall, the results partially confirm the study’s

original hypothesis of SROs’ negative impact on school connectedness and positive impact on

feelings of safety because while there was only a slight positive relationship between SRO

opinions and feelings of safety, experienced violence was more impactful and significant. By

investigating these topics of school safety and connectedness in a student population that differs

ethnically and socioeconomically than previous research, a more diverse picture is captured of

student perceptions of SROs and their effects on safety perceptions and school connectedness.

Certain limitations to the present study may have contributed to the partial confirmation

of the study’s hypothesis. Only a small proportion of students (6%) indicated having any kind of

interaction with their SRO, so it may have led to the insignificant relationship between

interactions and feelings of safety. Additionally, this study surveyed a sample of 100 students,

so it is too small of a sample population to generalize to the much larger student body of 4,480

students. Despite its limitations, this study’s results contribute valuable insight into how student

opinions of SROs as well as their experiences at school affect their feelings of safety and school

connectedness. According the Benjamin Kutsyuruba and colleagues, a positive school climate,

safe school environment, and well-being of students are integral to meeting their academic,

emotional, and social needs (Kutsyuruba et al. 2015). By investigating the effects of safety

measures, such as School Resource Officers, researcher can further their understanding of how

their implementation can affect students experiences in school. The present study has provided a

new understanding of these effects of SROs on students of a new and diverse population that can

be applied to an extensive part of the student experience.

Bibliography

Bosworth, K., Ford, L., & Hernandaz, D. (2011). School Climate Factors Contributing to Student and

Faculty Perceptions of Safety in Select Arizona Schools*. Journal of School Health,81(4), 194-

201.

Brown, B. (2006). Controlling Crime and Delinquency in the Schools. Journal of School Violence,4(4),

105-125.

Brown, B., & Benedict, W. R. (2005). Classroom Cops, What Do the Students Think? A Case Study of

Student Perceptions of School Police and Security Officers Conducted in an Hispanic

Community. International Journal of Police Science & Management, 7(4), 264-285.

Corley, C. (2018). Do Police Officers In Schools Really Make Them Safer? Retrieved from

https://www.npr.org/2018/03/08/591753884/do-police-officers-in-schools-really-make-them-

safer

Cornell, D. G., & Loper, A. B. (1998). Assessment of violence and other high-risk behaviors with a

school survey. School Psychology Review,27(2), 317-331.

Goldstein, S. E., Young, A., & Boyd, C. (2007). Relational Aggression at School: Associations with

School Safety and Social Climate. Journal of Youth and Adolescence,37(6), 641-654.

Jackson, A. (2002). Police‐school resource officers’ and students’ perception of the police and

offending. Policing: An International Journal of Police Strategies & Management,25(3), 631-

650.

Kutsyuruba, B., Klinger, D. A., & Hussain, A. (2015). Context and Implications Document for:

Relationships among school climate, school safety, and student achievement and well-being: A

review of the literature. Review of Education,3(2), 136-137.

McDevitt, J., & Panniello, J. (2005). National Assessment of School Resource Officer Programs: Survey

of Students in Three Large New SRO Programs. U.S. Department of Justice.

Rosenblatt, J. A., & Furlong, M. J. (1997). Assessing the Reliability and Validity of Student Self-

Reports of Campus Violence. Journal of Youth and Adolescence,26(2), 187-202.

Skiba, R., Simmons, A. B., Peterson, R., Mckelvey, J., Forde, S., & Gallini, S. (2004). Beyond Guns,

Drugs and Gangs. Journal of School Violence,3(2-3), 149-171.

Theriot, M. T. (2016). The Impact of School Resource Officer Interaction on Students’ Feelings About

School and School Police. Crime & Delinquency,62(4), 446-469.

Theriot, M. T., & Orme, J. G. (2014). School Resource Officers and Students’ Feelings of Safety at

School. Youth Violence and Juvenile Justice,14(2), 130-146.

Varjas, K., Henrich, C. C., & Meyers, J. (2009). Urban Middle School Students Perceptions of Bullying,

Cyberbullying, and School Safety. Journal of School Violence, 8(2), 159-176.

Addendum

My December 2018 GHC Study

the effects of school resource officers on feelings of

Documents